Abstract: In order to improve the accuracy and reliability of coal and gas outburst prediction results in rock cross-cut uncovering,and make up for the limitations of “spot” forecasting technology,taking X10903 working face of west well in Dawan Coal Mine as the test site,by analyzing the time spectrum and spectrum of acoustic emission monitoring signal and studying the variation rule of acoustic emission characteristic parameters,the application of acoustic emission prediction technology in rock cross-cut uncovering during roadway tunneling was investigated.The results show that the process of installing the sensor at the bottom of the hole can effectively isolate the interference signal near the installation position,based on the discrimination of frequency distribution,waveform and wavelength,the acoustic emission monitoring signal can be effectively filtered;the coal and gas outburst prediction results based on acoustic emission characteristic parameters in the process of rock cross-cut uncovering are consistent with the actual,moreover,the monitoring signal is consistent with the index of coal and gas outburst prediction in working face and the gas concentration after blasting,it is indicated that acoustic emission prediction technology can be applied to gas outburst prediction results in rock cross-cut uncovering,and provides a new auxiliary method for coal and gas outburst prediction in rock cross-cut uncovering of production mine.